DEV Community

#llm

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
I Packed DeepSeek V4 + Claude Code Into a Starter Kit. Clone It and Ship.

I Packed DeepSeek V4 + Claude Code Into a Starter Kit. Clone It and Ship.

Comments
3 min read
Deploying Agents: Containers, Orchestration, and Scaling the Loop

Deploying Agents: Containers, Orchestration, and Scaling the Loop

1
Comments 1
7 min read
Opus 4.8 barely moved the leaderboard. It moved the one number that decides if your agents can be trusted.

Opus 4.8 barely moved the leaderboard. It moved the one number that decides if your agents can be trusted.

Comments
5 min read
Fine-tuning — Domain-Specializing Models with LoRA

Fine-tuning — Domain-Specializing Models with LoRA

1
Comments
7 min read
Picking an Agent Framework in 2026: An Honest Verdict on Six of Them

Picking an Agent Framework in 2026: An Honest Verdict on Six of Them

Comments
6 min read
Silent Drift in Agent Decision Quality: Catching It Before Your Users Do

Silent Drift in Agent Decision Quality: Catching It Before Your Users Do

Comments
7 min read
Multi-Agent Coordination: Message-Bus Patterns That Keep Agents Sane

Multi-Agent Coordination: Message-Bus Patterns That Keep Agents Sane

Comments
6 min read
Six Bugs Only a Live Model Could Teach Us

Six Bugs Only a Live Model Could Teach Us

1
Comments 2
6 min read
Pydantic AI: Typed, Testable Agents for Engineers Who Like Guarantees

Pydantic AI: Typed, Testable Agents for Engineers Who Like Guarantees

Comments
6 min read
Decoding AI #1: Breaking LLMs with Obfuscated Log Malice (DeepSeek vs. Qwen)

Decoding AI #1: Breaking LLMs with Obfuscated Log Malice (DeepSeek vs. Qwen)

1
Comments
4 min read
Handoff-Driven Development

Handoff-Driven Development

Comments 1
9 min read
Hallucinations Are Not Always Wrong Facts: Sometimes They're Wrong Interpretations

Hallucinations Are Not Always Wrong Facts: Sometimes They're Wrong Interpretations

1
Comments 1
2 min read
Opus 4.8, Qwen, DeepSeek, and a Claude Code Failure: What I Could Actually Reproduce

Opus 4.8, Qwen, DeepSeek, and a Claude Code Failure: What I Could Actually Reproduce

Comments
3 min read
AI Basics: Key Concepts Every Software Engineer Should Know

AI Basics: Key Concepts Every Software Engineer Should Know

Comments
18 min read
I Ditched Vector Search for My Coding Agent's Memory. FTS5 Won.

I Ditched Vector Search for My Coding Agent's Memory. FTS5 Won.

Comments 1
4 min read
đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.